Flexible Neural Trees for Online Hand Gesture Recognition using surface Electromyography

Normal hand gesture recognition methods using surface Electromyography (sEMG) signals require designers to use digital signal processing hardware or ensemble methods as tools to solve real time hand gesture classification. Some methods could also result in complicated computational models, complex circuit connection and lower online recognition rate. It is therefore imperative to have good methods to explore a more suitable online design choice, which can avoid the problems mentioned above. An online hand gesture recognition model by using Flexible Neural Trees (FNT) and based on sEMG signals is proposed in this paper. The sEMG is a non-invasive, easy to record signal of superficial muscles from the skin surface, which has been applied in many fields of treatment and rehabilitation. The FNT model is generated and evolved based on the pre-defined simple instruction sets, which can solve highly structure dependent problem of the Artificial Neural Network (ANN). FNT method avoids complicated computation and inconvenience of circuit connection and also has an higher online recognition rate. Testing has been conducted using several continuous experiments conducted with five participants. The results indicate that the model is able to classify six different hand gestures up to 97.46% accuracy in real time.